Skip to main content

SELF ORGANISED SOCIETIES OF CONNECTIONIST INTELLIGENT AGENTS CAPABLE OF LEARNING

Objective

The current proposal aims to design communities of cooperating autonomous agents for maintenance missions in complex micro-fluidic environments such as those found in current and emergent platforms of artificial organs (e.g. artificial kidney dialysers). To accomplish this task the project adopts an integrated approach that will make use of principles of self-organization found in societies of social insects. Based on these principles the mission is accomplished by the emergent behaviour of colonies of simple micro-scale robotic agents. Novel, micro-scale gate evolvable spiking neural network architectures built specifically for the project will permit real time intelligent behaviour at the individual and social level. Macro-scale hybrid (software, hardware) and micro-scale (hardware) agents will be built for exploration of simulated and actual microfluidic environments respectively. The current proposal aims to design communities of cooperating autonomous agents for maintenance missions in complex micro-fluidic environments such as those found in current and emergent platforms of artificial organs (e.g. artificial kidney dialysers). To accomplish this task the project adopts an integrated approach that will make use of principles of self-organization found in societies of social insects. Based on these principles the mission is accomplished by the emergent behaviour of colonies of simple micro-scale robotic agents. Novel, micro-scale gate evolvable spiking neural network architectures built specifically for the project will permit real time intelligent behaviour at the individual and social level. Macro-scale hybrid (software, hardware) and micro-scale (hardware) agents will be built for exploration of simulated and actual microfluidic environments respectively.

OBJECTIVES
This project aims to investigate methods for engineering emergent collective behaviour in large societies of micro-nano scale collaborative autonomous agents (C.A.A.) that can learn and evolve. The applications objective for this project is to provide platforms for on-line maintenance and repair of filters for organ replacement therapy systems.

The project will:
(a) design a novel agent architectural design, which will serve as a blueprint for agent manufacture;
(b) implement an accompanying development environment, which will facilitate the evolution of a hierarchy of complex agent architectures;
(c) develop hardware technology platforms and processes encompassing high density interconnect (HDI) solutions to create tangible micro-nano agents and;
(d) implement a proof-of-concept system using micro-channel networks to replicate aspects of the priority application in order to investigate in-situ behaviour of the tangible agents.

DESCRIPTION OF WORK
In this proposal we are trying to solve a particularly difficult problem e.g. fault localization and repair in inaccessible micro-environment by using teams of very simple agents. The mission can be accomplished only if the societies of agents employed can manifest emergent intelligent collective behaviour. Such behaviour naturally arises in the context of stigmergic communication schemes, which are adopted in the project. Individual agents will generally be composed of:
- A sensor/actuator subsystem which will provide multiple input data and enable real-time navigation;
- A communication subsystem that will support self-organization of behaviour at a social level;
- A computational subsystem which will analyse the information of the sensory channels, control the actuator subsystems of the agent and initiate and maintain goal-directed individual and social behaviour among agents. The computational subsystem is composed of simple connectionist spiking neural networks. The latter are realized in gate evolvable spiking neural hardware, chosen for its ability to meet the real time signal processing imposed by the application context.

The project will demonstrate the feasibility of the concept by developing an early prototype; the project progress towards the ultimate goals will be measured by implementing and testing successive generations of agents. In accordance with potential application areas, two types of agents will be implemented and tested. - Macro-scale, exploring micro-environments in simulations.- Micro-scale containing reduced versions of the architectures for fault location and repair in actual microfluidic environments. An accompanying development environment will allow rapid prototyping and evaluation of successive generations of agents.

This environment will include modules that:
a) automatically generate agent architectures and;
b) evolve optimal architectures based on the team's performance in a micro-fluidic environment.

Funding Scheme

CSC - Cost-sharing contracts

Coordinator

UNIVERSITY OF PATRAS
Address
Rion Patras
26500 Patras
Greece

Participants (3)

RESEARCH ACADEMIC COMPUTER TECHNOLOGY INSTITUTE
Greece
Address
61, Riga Feraiou Street
26221 Patras
UNIVERSITY COLLEGE CORK - NATIONAL UNIVERSITY OF IRELAND, CORK
Ireland
Address
Western Road
Cork
UNIVERSITY OF ESSEX
United Kingdom
Address
Wivenhoe Park
CO4 3SQ Colchester